See also: IRC log
not sure why i'm so noisy...nothing at my end
i'll close some windows etc and see if that makes a difference
<David_> test
<scribe> Scribe: patrick_h_lauke
<Kathy> https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Proposed_Speech_Input
[discussion on whether we want to reopen the discussion about px and target sizes, decision to stick with what we decided last week as the value is likely to be discussed further once it reaches WCAG WG]
<Kim> https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Proposed_Single_Key_Shortcut_Alternative
<Kim> https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Proposed_Speech_Command_Mechanism
Kim: the first one connects speech input to keyboard access.
The second relates to turning all single character shortcuts off
Last one related to user being able to enter a multi-character string as customised shortcut
Kim: let's start with the first
Detlev: key question is what's impact on web content, what authors should or shouldn't do. it's not clear to me
i'm not sure if macros etc are done at web content level, or at AT level
or UA level
<David_> https://www.w3.org/TR/UNDERSTANDING-WCAG20/conformance.html#accessibility-supporteddef
<David_> https://www.w3.org/TR/UNDERSTANDING-WCAG20/conformance.html
Detlev: we should have examples of what can go wrong
Kim: if you grab keycodes before they can get to the program (user agent)
like Google Docs takes keycodes before speech recognition programme can interpret them
but we need more examples
Detlev: that would be a good example
the key thing is that it's operable through keyboard
that's main problem
Kathy: there is known issue when you do error checking, and that checking is done by capturing keystrokes rather than looking at value of input, the value is not known for instance because of how Dragon inputs text
<David_> https://www.w3.org/TR/2008/REC-WCAG20-20081211/#atdef
results in users not being able to continue
Kim: is this Dragon's fault?
Kathy: it's a Dragon issue, but also affects autocomplete in browser
as values are injected, rather than sending keycodes
Kim: what i do is have macros that are actual keystrokes
but should Dragon fix it?
Kathy: it's still an issue for autocomplete, might be other tools
Kim: is this related to google docs (it grabs all keystrokes)
Patrick: this SC seems contradictory with the other two SCs
it seems to say, essentially from our discussion: don't intercept keystrokes
<Detlev> Patrick: First SC may be redundant siince we require all to be operably by KB already
while the next two SCs talk about being able to turn them off, AND having shortcuts that are multi-chars
[longer discussion, also pointing to M6 that covers the validation issue mentioned by Kathy]
Patrick: propose that this first SC is in fact redundant (also as keyboard access itself is already in WCAG)
<David_> https://www.w3.org/TR/UNDERSTANDING-WCAG20/conformance.html
so drop this SC, and the more specific examples/situations just discussed are covered by Kim's other two SCs and my M6
David: agree this seems to be covered
but examples could be moved to those other cases
Kim: agree
so moving to the other two SCs
https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Proposed_Single_Key_Shortcut_Alternative
<Detlev> Patrick: Agrees that it may be coverd elsewhere - fals more to the realm of wah tthe AT should do (haviong different modes such as edit mode for text input)
<Detlev> Patrick (talking about the single key shortcuts)
<Detlev> Patrick: Pointer operable might cover alternatives to kb shortcuts
Kim: as these are SHORTcuts rather than only way to do things, turning them off shouldn't be a problem
Patrick: my proposed "Pointer" SC should take care of shortcuts not being the ONLY way to trigger something
[Kim explains how single-key shortcuts on Google Docs create problems when speech AT is on and user inadvertedly speaks, not knowing that speech AT is on and will fire commands to Google Docs]
Alan: sentence needs clarification
Patrick: is this not a problem with Dragon/AT not having clear "listen/don't listen" mode?
Kim: no modes are not efficient
<Detlev> Patrick: Can a mechanism to turn off/on AT resolve authors from the obligation to build turning off singlwe key shortcuts into their content?
<Detlev> Kim: No
<Detlev> Patrick: Information on modes and speech use micing commands and text input should go into benefits section
[Kim further explains about use of speech input]
Patrick: this would be good in benefits and in long run "Understanding..."
David: more changes are editorial, should we go forward
<jeanne> +1
<marcjohlic> +1 on going fwd with it
Kim: so do we go forward?
<chriscm> +1
David: +1
<Henny> +1
<Alan_Smith> +1
<Kathy> +1
Patrick: cautious +1
<Jatin_> +1
but we need techniques
<Detlev> +1
<Kim> +1
https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Proposed_Speech_Command_Mechanism
<Kim> https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Proposed_Speech_Command_Mechanism
Kim: moving to the third one, which is now the second one (since we removed first SC)
David: you want your shortnames being short not more than 3 or so words
<Detlev> Patrcik: Sites wanting to meet ths AAA SC would have to do a lot of work (settings section for defining commands etc) - the amount of work is quite large
<Detlev> Patrick: Multi character shortcuts are not easy to handle, you have to buffer input to look at the sequence - a lot harder than single key shortcuts
<Alan_Smith> +1 for AAA for this one
Kathy: we may also want to align with customisation/personalisation work happening
in COGA
Kim: want me to send message to Lisa to look at this?
Kathy: yes. it might roll into theirs, or have a separate AAA, but we need to by in sync
Kim: does explanation need beefing up as well?
Kathy: yes
Kim: I would prefer at AA rather than AAA, but i hear Patrick's concern and rather have this in
hopefully in future this will be more pervasive as an option, and may well be built into a higher level (UA, AT)
So I think i have my things to do for these
RESOLUTION: remove first SC, beef up second and third SC
<Kathy> https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/WCAG_2.1_Success_Criteria_Submission_Requirements
Kim: do you mean beef up 2nd/3rd SC and then we discuss it further?
David: no these are more editorial, TF is happy in principle
RESOLUTION: remove first SC, make further editorial changes to 2nd/3rd SC, send 3rd to COGA for sync
<David_> I think we need to docment that we have accepted 2, 3 to be submitted to the WG.
https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/WCAG_2.1_Success_Criteria_Submission_Requirements
<Detlev> Patrick: M1 (Pointer) and M3 (Pointer gestures) should be ready for discussion next week
<Detlev> https://github.com/detlevhfischer/Mobile-A11y-Extension/blob/patch-1/SCs/m1.md
<Detlev> https://github.com/detlevhfischer/Mobile-A11y-Extension/blob/patch-1/SCs/m3.md
<Detlev> damm wrong URL
Patrick: also happy to put things into github for others
Kathy: those marked with "x" (Reviewed by taskforce) can be put into the SC folder too
Patrick: I'll get those in
This is scribe.perl Revision: 1.144 of Date: 2015/11/17 08:39:34 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Found Scribe: patrick_h_lauke Inferring ScribeNick: patrick_h_lauke Present: Kathy Kim Alan patrick_h_lauke Detlev 1 Jatin marcjohlic WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 08 Sep 2016 Guessing minutes URL: http://www.w3.org/2016/09/08-mobile-a11y-minutes.html People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]